结合小波变换对心电信号中RR间期检测的一种新方法  

A new detection method for R-R intervals of electrocardiograph signals with wavelet transform

作　　者：袁静[1] 和卫星[1] 邓雪云[2] 吉奕[1] 吴秋明[1] 

机构地区：[1]江苏大学电气信息工程学院,江苏省镇江市212013 [2]南京航空航天大学民航学院,江苏省南京市210016

出　　处：《中国临床康复》2006年第13期116-118,共3页Chinese Journal of Clinical Rehabilitation

摘　　要：目的:探讨改善心电信号噪声干扰的方法,提高对心率变异信号中的RR间期检测精度。方法:①采用国际上通用的MIT/BIH数据库作为研究心率变异信号中RR间期的研究对象,选取其中的代表性数据组进行实验分析(受严重基线漂移影响的正常心电信号101.dat,大量室性期前收缩的心电信号105.dat,无噪声干扰的正常心电信号213.dat,心动过速的心电信号217.dat)。②利用小波变化方法结合Mallat算法对含有噪声的心电信号进行多尺度分析和信号重构,并利用R波的自身波形特点,采用相对周期极大值法来进行R波位置检测,进而计算RR间期序列值。结果:利用小波变换对含有噪声的信号进行噪声消除可以达到在很大程度保留原始信号的波形特征的同时又取得良好消噪效果的目的。能够显著减小工频干扰、基线漂移和肌电干扰等噪声对判别的影响。通过MIT/BIH数据库中四组有代表性特征的心电信号进行研究,发现采用相对周期极大值检测法可以显著减少检测中易出现的漏检和误检现象,快速而准确的获得RR间期的序列值。结论:小波变换法能够显著减少噪声对信号的干扰,特别是离散小波的应用使数字信号的处理由理论走向实际,结合Mallat快速算法,使得小波变换完全走向实用化。周期极大值法对心率变异信号中RR间期的检测有较好的精确度和快速性。AIM： To investigate the method of decreasing the noise disturbance in electrocardiosignal and improve the test precision of RR interval in heart rate variability （HRV） signals. METHODS： ①The MIT/BIH databases which was used aboard in international were taken as the investigate object to deal with the R-R intervals in HRV signals and experiment analysis was conducted with the selected representative dates.[Normal sinus rhythms interfered by baseline excursion 101.dat, ventricular fibrillation （VF） 105.dat, normal sinus rhythms （NSR） 213.dat, ventriculartachycardia （VT） 217.dat]. ②The wavelet transform was used to deal with the noised ECG signals by using the Mallat algorithm to decompose the signal into multi-scales and reconstruct it on selected scales. Then by the detection method of changed terms combined the characteristic of the value of R wave relative amplitude was detected so as to calculate the serial values of RR intervals. RESULTS： The result of wavelet transform used in denoising the signals was approximated the expectation value by decrease the noises such as power-line interference, the base-line interference, the electromyography （EMG） interference, etc., simultaneity, the wavelet transform could reserve the characteristics of the original signals in great degree. By investigating the four groups of characteristic data selected from the MIT/BIH database. It found that the detection method of changed terms combined the characteristics of the value of R wave could reduce the phenomena of misschecks The RR intervals could be gotten exactly and quickly. CONCLUSION： The wavelet transform can reduce the noising interference distinctly. Especially, the discrete wavelet transform make the digital signals analyses from theory into practices by using the Mallat algorithm; the detection method of changed terms combined the characteristic of the value of R wave can make the detection more accuracy and more quickly.

关 键 词：心电描记术 心率 心脏/生理学 

分 类 号：R319[医药卫生—基础医学]